PSI - Issue 47

Umberto De Maio et al. / Procedia Structural Integrity 47 (2023) 469–477 Author name / Structural Integrity Procedia 00 (2019) 000–000

470

2

1. Introduction In the last years, the use of fiber-reinforced polymer (FRP) composites as a strengthening system for existing structures has become very popular due to their ability to increase strength without significantly changing the stiffness of the structural element. As is well-known, however, attention must be paid to the possible failure mechanisms that can occur, one among the others the debonding phenomena (Chen et al., 2011; Sebastian, 2001). In addition, more complex brittle failures in such kinds of materials can be induced also by the onset of instabilities commonly observed due to the adoption of reinforcing fibers acting at different structural scales based on advanced microstructures (De Maio et al., 2023b; Maio et al., 2020, 2020; Pranno et al., 2022a). Such failure mechanisms could negatively affect the global behavior of the structure, reducing its load-bearing capacity (Aram et al., 2008; De Maio et al., 2019a). To reduce the risks of such phenomena occurring, Structural Health Monitoring (SHM) procedures are useful tools to assess damage detection and its location within existing structures (Yang and Yang, 2018). They are based on the collection of periodic measurements obtained from sensors installed on structures, in order to track the changes that damage brings on the static and dynamic properties of a structural system (Frangopol and Curley, 1987; Lynch, 2006). Appropriate mathematical models are needed to transform the data obtained during the monitoring phases into damage indicators, to have useful information for understanding the health status of the structures. As a matter of fact, variations in the properties of structures, especially the dynamic ones (natural vibration frequencies and vibration mode shapes), can be used to recognize the presence of damage, its location, and the level of risk to which that structure is exposed (Abdel Wahab and De Roeck, 1999; Fugate et al., 2001; Greco et al., 2018; Salawu, 1997; Zhu and Zhang, 2023). Several approaches for damage detection have been proposed in the technical literature, mainly grouped into data-based and model-based methods (Farrar et al., 2001; Karbhari and Lee, 2009). The former use so-called global quantities, based on static and dynamic parameters obtained during monitoring phases, and compare them with the values of the undamaged structures to identify the presence, location, and magnitude of the damage. On the other hand, the latter classes are based on updating finite element models (FEMs); a set of parameters belonging to a preliminary model representing the initial conditions, i.e. the monitored and undamaged structure, are opportunely modified to achieve a match with the actual conditions of the structure subjected to damage phenomena (Doebling et al., 1998; Magalhães et al., 2012). In particular, an optimization problem is developed to reduce the differences between the experimentally measured dynamic response and the numerical model. By achieving an optimal match, the location and magnitude of the damage can be detected. In the field of model-based techniques, different numerical methods have been introduced in the past years based on discrete or smeared fracture approaches (Hanif et al., 2018). Methods belonging to the discrete fracture approach, including cohesive models, have the main advantage of appropriately predicting the crack pattern induced by the applied loads, as well as the complex damage phenomena typical of reinforced concrete (RC) structures. Moreover, they are versatile and applicable to different types of quasi-brittle materials, such as masonry (Conde et al., 2017), concrete elements (De Maio et al., 2021, 2022a), and RC elements enhanced with embedded nanomaterials (Ammendolea et al., 2023), combining reliability, in terms of expected results, and lower computational costs. On the other hand, methods belonging to smeared fracture approach can provide a more accurate response in terms of the load-displacement curve despite less accuracy in reproducing the crack path. However, a damage identification method for the analysis of strengthened RC structures, whether based on smeared or discrete fracture approach, must include an accurate numerical model (for instance based on multiscale approaches (Greco et al., 2020a, 2020b) or advanced finite element approach (Ammendolea et al., 2021; Greco et al., 2021; Pascuzzo et al., 2022)) in conjunction with an appropriate monitoring campaign, in order to better analyze the behavior of existing structures and all complex collapse mechanisms that may occur (Mansouri et al., 2015). In this perspective, within the structural health monitoring procedures based on the coupling between numerical models and experimental data, a numerical model has been developed to investigate the crack-induced degradation of vibration characteristics in reinforced concrete (RC) plated elements. In order to take into account all complex nonlinear phenomena, such as concrete crushing, rebar yielding and FRP debonding, an extension of the cohesive model proposed by some of the authors in (De Maio et al., 2022b) has been employed in a 2D finite element framework. In particular, a cohesive zone model, used to predict the crack onset and propagation in the concrete phase and the debonding phenomena at the interface between concrete and FRP reinforcement system, has been used in conjunction with an embedded truss model (ETM) able to reproduce the mechanical interaction between concrete and steel rebars.